3 ms·

I think this is a neat idea and has some valuable information in it, except for the images. I've been working with film for a long time. I've shot a ridiculous

I think this is a neat idea and has some valuable information in it, except for the images. I've been working with film for a long time. I've shot a ridiculous number of films from 35mm to 4x5, scanned them on everything from roll-film scanners to Flextight to drum scanners, and printed on many different papers. My conclusion after going really deep down this rabbit hole is that, except for a very small number of films with really unique characteristics (most of them no longer made), you just can't tell much of anything from a digital image because of the lack of standardization across: A) Setting/Exposure B) Film Developer C) Scanner/calibration/profile D) Whether the image is a film scan or a print scan

rylax 7y ago
I waited for this comment for quite a while, hah. Yes, I agree with you on those assumptions to the most extent! It is really difficult to find out the exact characteristics given only the photos since the outcome can be altered by so many factors. However, even though I was aware of this - I still tended to look at example photos on tumblr because at least from my experience the sum of all viewed images do give you a good overview imho. The reason why I handpicked the first six photos by hand is actually because of this. When picking the images I tried to pick the ones that most match with the given characteristic. Off course it is still a very subjective matter.
- staticautomatic 7y ago
  I think at the end of the day, the right conclusion is that an image is exemplary of the process used to create it. If you agree with that, then I hope you'll consider including information about the process along with the image. Without that context, I don't think you can reasonably attribute a "look" to the film itself. Minimally, I think you'd need to know: A) What's the film format/size? B) Was it shot and processed normally? (i.e. not pushed, pulled, cross-processed, or developed with a staining developer) C) Is this image a scan of a negative or a scan of a print?
  - rylax 7y ago
    Well, while this approach would make it much clearer it is unfortunately also very unfeasible to do. Neither a photo api would give those details nor would it be possible to aggregate all those information by hand for each film. Thanks for your input!

rylax 7y ago

I waited for this comment for quite a while, hah. Yes, I agree with you on those assumptions to the most extent! It is really difficult to find out the exact characteristics given only the photos since the outcome can be altered by so many factors. However, even though I was aware of this - I still tended to look at example photos on tumblr because at least from my experience the sum of all viewed images do give you a good overview imho. The reason why I handpicked the first six photos by hand is actually because of this. When picking the images I tried to pick the ones that most match with the given characteristic. Off course it is still a very subjective matter.

staticautomatic 7y ago
I think at the end of the day, the right conclusion is that an image is exemplary of the process used to create it. If you agree with that, then I hope you'll consider including information about the process along with the image. Without that context, I don't think you can reasonably attribute a "look" to the film itself. Minimally, I think you'd need to know: A) What's the film format/size? B) Was it shot and processed normally? (i.e. not pushed, pulled, cross-processed, or developed with a staining developer) C) Is this image a scan of a negative or a scan of a print?
- rylax 7y ago
  Well, while this approach would make it much clearer it is unfortunately also very unfeasible to do. Neither a photo api would give those details nor would it be possible to aggregate all those information by hand for each film. Thanks for your input!